- Email: [email protected]
SL 82.0715, an NMDA antagonist acting at the polyamine site, does not induce neurotoxic effects on rat cortical neurons
Neuroscience Letters, 137 (1992) 193-197 © 1992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 0304-3940/92/$ 05.00
193
NSL 08493
SL 82.0715, an N M D A antagonist acting at the polyamine site, does not induce neurotoxic effects on rat cortical neurons D. Duval a, N. Roome b, C. Gauffeny b, J.P. Nowicki a and B. Scatton a Synthelabo Recherche (L.E.R.S.), aBiology Department, Bagneux (France) and bDepartment of Drug Safety, Gargenville (France) (Received 1 October 1991; Revised version received 27 December 1991; Accepted 27 December 1991)
Key words: Neurotoxicity; N-Methyl-D-aspartate receptor antagonist; Polyamine site In the present study, we have examined by light and electron microscopy whether SL 82.0715, a polyamine site-directed N-methyl-D-aspartate (NMDA) antagonist, causes pathological changes in cerebrocortical neurons similar to those observed with NMDA receptor channel blockers in the rat brain. Dizocilpine (1, 2 and 5 mg. kg-1, s.c.) induced a dose-dependent vacuolization of the neuronal cytoplasm in specific neurons of the retrosplenial and posterior cingulate cortices (layers III and IV) even at the lowest dose studied, at 6 h post-injection. In contrast, SL 82.0715 (10 and 30 mg. kg-1 i.p., 6 h post-injection) did not induce such morphological alterations. These results indicate that NMDA receptor blockade is not necessarily associated with alterations of cortical neuronal morphology.
In the literature, there is a growing body of evidence for the neuroprotective efficacy of all types of N-methylD-aspartate (NMDA) receptor antagonists in experimental models of focal cerebral ischaemia in various species [1, 10, 13, 18] and against acute neuronal injury induced by spinal cord or brain trauma [7, 9]. There are, however, concerns about the potential adverse effects in the CNS of NMDA receptor antagonists, particularly those that act at the channel site. Thus, NMDA receptor channel blockers, e.g., phencyclidine and dizocilpine, are behavioral stimulants in animals [3, 22] and phencyclidine is psychotomimetic in man. Moreover, in drug discrimination experiments in the rat, dizocilpine generalizes to the phencyclidine cue [12, 23]. These drugs also disrupt learning and memory in the rodent and in the monkey [8, 23]. However, this profile of unwanted effects is not shared by all NMDA receptor antagonists [17, 20, 22, 23]. Another serious problem with NMDA receptor antagonists is their apparent neurotoxic effects in the rat [16]. Relatively low doses of phencyclidine or dizocilpine have been found to produce a vacuolization in multipolar and pyramidal-shaped neurons in layers III and IV of the posterior cingulate and retrosplenial cortices in the rat. These changes subside within 18 to 24 h after acute drug administration and are no longer seen after Correspondence: B. Scatton, Synthelabo Recherche, Biology Department, 31 avenue Paul Vaillant-Couturier, 92220 Bagneux, France.
prolonged treatment. However, while vacuolization is reversible after low doses of these drugs, higher doses (e.g. 5 mg.kg -1 of dizocilpine) are associated with the appearance of occasional necrotic neurons 48 h after treatment [2]. SL 82.0715 is a potent non-competitive NMDA receptor antagonist that acts at the polyamine modulatory site [4, 5]. In common with other NMDA receptor antagonists, SL 82.0715 displays neuroprotective efficacy in experimental models of focal cerebral ischaemia in various species [10, 18] and in a model of brain trauma caused by fluid percussion in the rat [21]. However, SL 82.0715 does not possess the undesirable side effects observed with the NMDA receptor antagonists acting at the channel site. Thus, SL 82.0715 is not stimulant in the rodent, does not generalize to phencyclidine in animals [17] and is not psychotomimetic in man (EL. Morselli, personal communication). The aim of the present study was to investigate whether SL 82.0715 induces a vacuolar reaction in the cytoplasm of neurons similar to the one seen after the administration of drugs that interact with the NMDA receptor-gated channel. All experiments were performed on male SpragueDawley rats (300-350 g). SL 82.0715 was administered i.p. as a solution in 0.1 M tartaric acid at the doses of 10 m g - k g I (3 rats) and 30 mg.kg i (9 rats). Dizocilpine (MK- 801) was administered s.c. as an aqueous solution at the doses of 1 mg. kg-1 (3 rats), 2 mg. k g I (3 rats) and
::)
195
5 mg-kg- l (9 rats). Control rats (n = 8) received 0.1 M tartaric acid. Two rats in the SL 82.0715 (30 mg.kg -~) and dizocilpine (5 mg. kg-~)-treated groups and two rats in the control group were randomly selected for the electron microscopic study. Six hours after drug administration, the rats were anaesthetized with pentobarbital and their brains were fixed in situ via an intracardiac perfusion of a fixative. For the histopathological evaluation by light microscopy, the brains were fixed with FAM (40% formaldehyde - - glacial acetic acid - - methanol, 1:1:8) and embedded in paraffin. Twenty semi-serial coronal sections (5 pm thick) were cut through the whole posterior cingulate and retrosplenial cortices and stained with hemalun and eosin. The sections were examined by two independent observers that were unaware of the treatment administered. For the electron microscopic study, the brains were perfused with 2.5% glutaraldehyde in cacodylate buffer (0.1 M, pH 7.2). Twelve small pieces (1 mm 3) from the retrosplenial and posterior cingulate cortices of each animal were taken, post-fixed in 1% osmium tetroxide and embedded in epon-araldite. Semi-thin sections (1 /.tm) were cut and stained with Azur blue for histopathological confirmation of the lesion by light microscopy. Ultra-thin sections (80 nm) were prepared and stained with uranyl acetate and lead citrate for ultrastructural evaluation. Examination by light microscopy of brain sections prepared from dizocilpine-treated rats revealed that neuropathological changes were present in the posterior cingulate and retrosplenial cortices in all rats administered with the doses of 5 mg- k g ~ and 2 mg. kg -~ and in 2 out of 3 animals treated with the dose of 1 mg- k g 1. In rats treated with the dose of 5 mg.kg ~, some neurons appeared hypertrophied with a swollen clear cytoplasm and an enlarged nucleus (Fig. lb). These changes were restricted to medium-sized and large neurons (multipolar and pyramidal shaped neurons) in layers III and sometimes IV of the posterior cingulate and retrosplenial cortices. The small neurons located in layer II never displayed such neuronal changes. The extent of the dizo-
cilpine-induced pathomorphological changes was dosedependent: while only few abnormal neurons could be detected at the lowest dose administered (1 mg.kg -1, s.c.), a large number of neurons were altered at the highest dose studied (5 mg. kg-1, s.c.). Electron microscopic examination of the affected neurons confirmed the presence of multiple vacuoles with heterogenous sizes in the cytoplasmic compartment (Fig. le). These vacuoles apparently originated both from the dilatation of the endoplasmic reticulum and from swollen mitochondria (the external double membrane and the altered cristae can be partially distinguished). Disaggregation of polyribosomes and loss of microtubules were also observed with the presence of phagolysosomes. In none of the SL 82.0715-treated rats, even at the highest dose administered (30 mg. kg t, i.p.), could such pathomorphological alterations be detected. By light microscopy, brain sections were indistinguishable from those of control rats (Fig. lc). All the neurons in the posterior cingulate and retrosplenial cortices displayed a normal morphology: they were angular in shape with a dense basophilic homogenous cytoplasm and a relatively large nucleus with an extended chromatin and an occasional prominent nucleolus. Light microscopic quantitative evaluation of neuronal lesions performed on 20 serial coronal sections (taken along the rostro-caudal extent of the posterior cingulate and retrosplenial cortices) and stained with hemalun and eosin showed that the cortex was totally devoid of lesioned neurons. The electron microscopic examination confirmed that SL 82.0715 did not produce any neuronal vacuolization in the posterior cingulate and retrosplenial cortices (Fig. 1f). The ultrastructure of the neurons was very similar to that found in control rats. Cytoplasmic organelles such as endoplasmic reticulum, polyribosomes, mitochondria, microtubules and Golgi apparatus could be easily distinguished and were not altered. As previously reported [2, 16], an important cytoplasmic vacuolization of some retrosplenial and posterior cingulate neurons was observed in the present study after
6-Fig. 1. Comparative effects of an acute administration of dizocilpine and SL 82.0715 on neuronal morphology in the rat posterior cingulate cortex. a: semi-thin section of normal neurons in a control animal in layer III of the posterior cingulate cortex (arrows). The cytoplasm of large bipolar or pyramidal neurons is well stained. Bar = 30/2m. b: semi-thin section of vacuolated neurons (arrows) in an animal treated with 5 m g . k g ~ s.c. of dizocilpine. The cytoplasm of the altered neurons appears swollen and pale with n u m e r o u s vacuoles. Bar = 30/2m. c: semi-thin section of unaltered neurons (arrows) in an animal treated with 30 m g . kg ~ i.p. of SL 82.0715. The cytoplasm of the neurons is normally well stained. Bar = 30/2m. d: electron micrograph of a neuron in a control animal showing mitochondria (M), endoplasmic reticulum (ER), microtubules (MT), polyribosomes (P) and Golgi apparatus (G). Bar = 0.6/tm. e: electron micrograph of a neuron in an animal treated with dizocilpine (5 m g . kg -~, s.c.) showing swollen mitochondria (arrows), dilated endoplasmic reticulum (arrowheads), and phagolysosome (curved arrow). Note the normal mitochondria (M) in the adjacent tissue. Bar = 0.6/.tm. f: electron micrograph from an animal treated with 30 m g . kg -t i.p. of SL 82.0715 showing normal ultrastructure. Bar = 0.6/lm.
196
acute administration of 5 mg. kg ~ of dizocilpine, a dose which represents 25 times its neuroprotective dose in a model of focal ischemia in the mouse [10]. Although only light microscopic evaluation has been performed at the lower dose of dizocilpine (1 mg. kg -~, i.e. 5 times the neuroprotective dose in the mouse), pathomorphological changes could also be evidenced in 2 out of 3 animals. Pathomorphological changes of cerebrocortical neurons have also been reported with phencyclidine and two other N M D A channel ligands, tiletamine and ketamine in the rat. The order of potency with which these agents induced this effect (dizocilpine > phencyclidine > tiletamine > ketamine) is the same as their order of affinity for binding to the N M D A receptor channel site [16]. Moreover, an acute neuronal vacuolization has been found after local injection of the competitive N M D A receptor antagonist, D-2-amino-5-phosphonopentanoate, directly into the cingulate cortical region [16]. In contrast, the present study demonstrates that the polyamine site-directed N M D A antagonist SL 82.0715, even when administered at a dose of 30 mg. kg ~which represents 27 times its neuroprotective dose (EDs0 = 1.1 mg/kg, i.p.) in a model of focal cerebral ischaemia in the mouse [10], did not induce any neuropathological alteration in the posterior cingulate and retrosplenial cortices in the rat. Similarly, L-687,414 (the (+)cis-methyl analogue of HA 966), an N M D A receptor antagonist acting at the glycine site, also failed to alter cortical neuronal morphology [11]. These results suggest that although all types of N M D A receptor antagonists produce a blockade of N M D A receptor function, they have markedly different effects on cerebrocortical neuronal morphology. Thus, the appearance of morphological alterations in cerebrocortical neurons in response to N M D A receptor antagonists may depend on the specific component of the N M D A receptor with which they interact. The reason why N M D A receptor channel blockers but not polyamine and glycine site antagonists induce neurotoxic effects is as yet unclear. Different conductance levels or states of N M D A receptor-gated calcium channels may be affected by these drugs resulting in a qualitatively different blockade of N M D A receptor function. Alternatively, it is possible that the pathomorphological effects of N M D A receptor channel blockers in some cortical areas is unrelated to their interaction with the N M D A receptor, but is due to an unknown property of these drugs. The mechanism whereby N M D A receptor channel antagonists induce neuronal vacuolization of cortical neurons remains unclear. The presence of early degenerative changes such as cytoplasmic vacuoles may be the structural consequence of excessive cell metabolism in this region due to the large increase in cerebral metabolic rate
measured after the administration of N M D A receptorgated channel blockers [14]. Furthermore, after administration of dizocilpine or ketamine, the increase in cerebral blood flow in the posterior cingulate cortex greatly exceeds the increase in metabolism, thus leading to an 'uncoupling' between flow and metabolism [6, 15]. It is not clear whether the increased metabolic rate occurs in the same cells that manifest pathomorphological changes, or in other neural elements in the same region, but it seems likely that the two phenomena are causally related. In support of this view, L-687,414 and SL 82.0715 which do not increase limbic glucose metabolism in the rat brain [11, 19] do not induce pathomorphological changes in cingulate and retrosplenial cortical neurons. In conclusion, these findings suggest that neuroprotection mediated through blockade of the N M D A receptor complex is not necessarily associated with abnormal cerebrocortical neuronal morphology. The presence or absence of pathomorphological alterations with N M D A receptor antagonists seems to be related to the specific site of the N M D A receptor with which they interact. 1 Albers, G.W., Goldberg, M.R and Choi, D.W., N-Methyi-Daspartate antagonists: ready for clinical trial in brain ischaemia? Ann. Neurol., 25 (1989) 398~,03. 2 Allen, H.L. and Iversen, L.L., Phencyclidine, dizocilpine and cerebrocortical neurons, Science, 247 (1990) 221. 3 Balster, R.L., The behavioral pharmacology of phencyclidine. In H.Y. Meltzer (Ed.), Psychopharmacology: the Third Generation of Progress, Raven, New York, 1987, pp. 1573-1579. 4 Carter, C.J., Benavides, J., Legendre, P., Vincent, J.D., Noel, F., Thuret, F., Lloyd, K.G., Arbilla, S., Zivkovic, B., MacKenzie, E.T., Scatton, B. and Langer, S.Z., Ifenprodil and SL 82.0715 as cerebral anti-ischemic agents. II. Evidence for N-methyl-D-aspartate receptor antagonist properties, J. Pharmacol. Exp. Ther., 247 (1988) 1222 1232. 5 Carter, C.J., Lloyd, K.G., Zivkovic, B. and Scatton, B., Ifenprodil and SL 82.0715 as cerebral anti-ischaemic agents. III. Evidence for antagonistic effects at the polyamine modulatory site within the NMDA receptor complex, J. Pharmacol. Exp. Ther., 253 (1990) 475 482. 6 Cavazzuti, M., Porro, C.A., Biral, G.P., Benassi, C. and Barbieri, G.C., Ketamine effects on local cerebral blood flow and metabolism in the rat, J. Cereb. Blood Flow Metab., 7 (1987) 806-811. 7 Faden, A.I., Demediuk, R, Panter, S.S. and Vink, R., The role of excitatory amino acids and NMDA receptors in traumatic brain injury, Science, 244 (1989) 798 800. 8 France, C.P., Moerschbaecber, J.M. and Woods, J.M., MK-801 and related compounds in monkeys: discriminative stimulus effects and effects on a conditional discrimination, J. Pharmacol. Exp. Ther., 257 (1991) 727-734. 8 Gomez-Pinilla, F., Tram, H., Cotman, C.W. and Nieto-Sampedro, M., Neuroprotective effect of MK-801 and U-50488 H after contusive spinal cord injury, Exp. Neurol., 104 (1989) 118-124. 10 Goni, B., Benavides, J., MacKenzie, E.T. and Scatton, B., The pharmacotherapy of focal cortical ischaemia in the mouse, Brain Res., 522 (1990) 290 307.
197 11 Hargreaves, R.J., Rigby, M., Smith, D., Foster, A., Hurley, C.J. and Hill, R.G., L 687,414(+)cis-4-methyl-HA-966, an NMDA receptor antagonist acting at the glycine site, does not alter glucose metabolism or neuronal morphology at neuroprotective dose levels, J. Cereb. Blood Flow Metab., 11 Suppl. 2 (1991) S 301. 12 Jackson, A. and Sanger, D.J., Is the discriminative stimulus produced by phencyclidine due to an interaction with N-methyl-Daspartate receptors? Psychopharmacology, 96 (1988) 87-92. 13 Meldrum, B., Protection against ischaemic neuronal damage by drug acting on excitatory neurotransmission, Cerebrovasc. Brain Metab. Rev., 2 (1990) 27-57. 14 Nehls, D.G., Kurumaji, A., Park, C.K. and McCulloch, J., Differential effects of competitive and non-competitive N-methyl-Daspartate antagonists on glucose use in the limbic system, Neurosci. Lett., 91 (1988) 204-210. 15 Nehls, D.G., Park, C.K., MacCormack, A.G. and McCulloch, J., The effects of N-methyl-D-aspartate receptor blockade with MK801 upon the relationship between cerebral blood flow and glucose utilisation, Brain Res., 511 (1990) 271-279. 16 Olney, J.W., Labruyere, J. and Price, M.T., Pathological changes induced in cerebrocortical neurons by phencyclidine and related drugs, Science, 244 (1989) 1360-1362. 17 Perrault, G., Morel, E., Joly, D., Sanger, D.J. and Zivkovic, B., Comparison of the pharmacological profiles of four NMDA antagonists: ifenprodil, SL 82.0715, MK-801 and CPP in mice, Br. J. Pharmacol., 97 (1988) 580 P.
18 Scatton, B., Carter, C., Benavid~s, J. and Giroux, C., N-Methyl-Daspartate receptor antagonists: a novel therapeutic perspective for the treatment of ischemic brain injury, Cerebrovasc. Dis., 1 (1991) 121-135. 19 Scatton, B., Cudennec, A., Duverger, D., MacKenzie, E.T., Nowicki, J.P. and Zivkovic, B., Effects of SL 82.0715, an NMDA receptor antagonist acting at the polyamine modulatory site, on local cerebral glucose use in the rat brain, J. Cereb. Blood Flow Metab., 11 Suppl. 2 (1991) S 312. 20 Singh, L., Menzies, R. and Tricklebank, M.D., The discriminative stimulus properties of (+)HA-966, an antagonist at the glycine/Nmethyl-D-aspartate receptor, Eur. J. Pharmacol., 186 (1990) 129132. 21 Toulmond, S., Serrano, A., Benavides, J. and Scatton, B., Neuroprotective effects of SL 82.0715 in an experimental model of brain trauma, J. Cereb. Blood Flow Metab., 11 Suppl. 2 (1991) S 313. 22 Tricklebank, M.D., Singh, L., Oles, R.J., Preston, C. and Iversen, S.D., The behavioral effects of MK-801: a comparison with antagonists acting non-competitively and competitively at the NMDA receptor, Eur. J. Pharmacol., 167 (1989) 127-135. 23 Willetts, J., Balster, R.L. and Leander, J.D., The behavioral pharmacology of NMDA receptor antagonists, Trends Neurosci., 11 (1990) 423428.
193
NSL 08493
SL 82.0715, an N M D A antagonist acting at the polyamine site, does not induce neurotoxic effects on rat cortical neurons D. Duval a, N. Roome b, C. Gauffeny b, J.P. Nowicki a and B. Scatton a Synthelabo Recherche (L.E.R.S.), aBiology Department, Bagneux (France) and bDepartment of Drug Safety, Gargenville (France) (Received 1 October 1991; Revised version received 27 December 1991; Accepted 27 December 1991)
Key words: Neurotoxicity; N-Methyl-D-aspartate receptor antagonist; Polyamine site In the present study, we have examined by light and electron microscopy whether SL 82.0715, a polyamine site-directed N-methyl-D-aspartate (NMDA) antagonist, causes pathological changes in cerebrocortical neurons similar to those observed with NMDA receptor channel blockers in the rat brain. Dizocilpine (1, 2 and 5 mg. kg-1, s.c.) induced a dose-dependent vacuolization of the neuronal cytoplasm in specific neurons of the retrosplenial and posterior cingulate cortices (layers III and IV) even at the lowest dose studied, at 6 h post-injection. In contrast, SL 82.0715 (10 and 30 mg. kg-1 i.p., 6 h post-injection) did not induce such morphological alterations. These results indicate that NMDA receptor blockade is not necessarily associated with alterations of cortical neuronal morphology.
In the literature, there is a growing body of evidence for the neuroprotective efficacy of all types of N-methylD-aspartate (NMDA) receptor antagonists in experimental models of focal cerebral ischaemia in various species [1, 10, 13, 18] and against acute neuronal injury induced by spinal cord or brain trauma [7, 9]. There are, however, concerns about the potential adverse effects in the CNS of NMDA receptor antagonists, particularly those that act at the channel site. Thus, NMDA receptor channel blockers, e.g., phencyclidine and dizocilpine, are behavioral stimulants in animals [3, 22] and phencyclidine is psychotomimetic in man. Moreover, in drug discrimination experiments in the rat, dizocilpine generalizes to the phencyclidine cue [12, 23]. These drugs also disrupt learning and memory in the rodent and in the monkey [8, 23]. However, this profile of unwanted effects is not shared by all NMDA receptor antagonists [17, 20, 22, 23]. Another serious problem with NMDA receptor antagonists is their apparent neurotoxic effects in the rat [16]. Relatively low doses of phencyclidine or dizocilpine have been found to produce a vacuolization in multipolar and pyramidal-shaped neurons in layers III and IV of the posterior cingulate and retrosplenial cortices in the rat. These changes subside within 18 to 24 h after acute drug administration and are no longer seen after Correspondence: B. Scatton, Synthelabo Recherche, Biology Department, 31 avenue Paul Vaillant-Couturier, 92220 Bagneux, France.
prolonged treatment. However, while vacuolization is reversible after low doses of these drugs, higher doses (e.g. 5 mg.kg -1 of dizocilpine) are associated with the appearance of occasional necrotic neurons 48 h after treatment [2]. SL 82.0715 is a potent non-competitive NMDA receptor antagonist that acts at the polyamine modulatory site [4, 5]. In common with other NMDA receptor antagonists, SL 82.0715 displays neuroprotective efficacy in experimental models of focal cerebral ischaemia in various species [10, 18] and in a model of brain trauma caused by fluid percussion in the rat [21]. However, SL 82.0715 does not possess the undesirable side effects observed with the NMDA receptor antagonists acting at the channel site. Thus, SL 82.0715 is not stimulant in the rodent, does not generalize to phencyclidine in animals [17] and is not psychotomimetic in man (EL. Morselli, personal communication). The aim of the present study was to investigate whether SL 82.0715 induces a vacuolar reaction in the cytoplasm of neurons similar to the one seen after the administration of drugs that interact with the NMDA receptor-gated channel. All experiments were performed on male SpragueDawley rats (300-350 g). SL 82.0715 was administered i.p. as a solution in 0.1 M tartaric acid at the doses of 10 m g - k g I (3 rats) and 30 mg.kg i (9 rats). Dizocilpine (MK- 801) was administered s.c. as an aqueous solution at the doses of 1 mg. kg-1 (3 rats), 2 mg. k g I (3 rats) and
::)
195
5 mg-kg- l (9 rats). Control rats (n = 8) received 0.1 M tartaric acid. Two rats in the SL 82.0715 (30 mg.kg -~) and dizocilpine (5 mg. kg-~)-treated groups and two rats in the control group were randomly selected for the electron microscopic study. Six hours after drug administration, the rats were anaesthetized with pentobarbital and their brains were fixed in situ via an intracardiac perfusion of a fixative. For the histopathological evaluation by light microscopy, the brains were fixed with FAM (40% formaldehyde - - glacial acetic acid - - methanol, 1:1:8) and embedded in paraffin. Twenty semi-serial coronal sections (5 pm thick) were cut through the whole posterior cingulate and retrosplenial cortices and stained with hemalun and eosin. The sections were examined by two independent observers that were unaware of the treatment administered. For the electron microscopic study, the brains were perfused with 2.5% glutaraldehyde in cacodylate buffer (0.1 M, pH 7.2). Twelve small pieces (1 mm 3) from the retrosplenial and posterior cingulate cortices of each animal were taken, post-fixed in 1% osmium tetroxide and embedded in epon-araldite. Semi-thin sections (1 /.tm) were cut and stained with Azur blue for histopathological confirmation of the lesion by light microscopy. Ultra-thin sections (80 nm) were prepared and stained with uranyl acetate and lead citrate for ultrastructural evaluation. Examination by light microscopy of brain sections prepared from dizocilpine-treated rats revealed that neuropathological changes were present in the posterior cingulate and retrosplenial cortices in all rats administered with the doses of 5 mg- k g ~ and 2 mg. kg -~ and in 2 out of 3 animals treated with the dose of 1 mg- k g 1. In rats treated with the dose of 5 mg.kg ~, some neurons appeared hypertrophied with a swollen clear cytoplasm and an enlarged nucleus (Fig. lb). These changes were restricted to medium-sized and large neurons (multipolar and pyramidal shaped neurons) in layers III and sometimes IV of the posterior cingulate and retrosplenial cortices. The small neurons located in layer II never displayed such neuronal changes. The extent of the dizo-
cilpine-induced pathomorphological changes was dosedependent: while only few abnormal neurons could be detected at the lowest dose administered (1 mg.kg -1, s.c.), a large number of neurons were altered at the highest dose studied (5 mg. kg-1, s.c.). Electron microscopic examination of the affected neurons confirmed the presence of multiple vacuoles with heterogenous sizes in the cytoplasmic compartment (Fig. le). These vacuoles apparently originated both from the dilatation of the endoplasmic reticulum and from swollen mitochondria (the external double membrane and the altered cristae can be partially distinguished). Disaggregation of polyribosomes and loss of microtubules were also observed with the presence of phagolysosomes. In none of the SL 82.0715-treated rats, even at the highest dose administered (30 mg. kg t, i.p.), could such pathomorphological alterations be detected. By light microscopy, brain sections were indistinguishable from those of control rats (Fig. lc). All the neurons in the posterior cingulate and retrosplenial cortices displayed a normal morphology: they were angular in shape with a dense basophilic homogenous cytoplasm and a relatively large nucleus with an extended chromatin and an occasional prominent nucleolus. Light microscopic quantitative evaluation of neuronal lesions performed on 20 serial coronal sections (taken along the rostro-caudal extent of the posterior cingulate and retrosplenial cortices) and stained with hemalun and eosin showed that the cortex was totally devoid of lesioned neurons. The electron microscopic examination confirmed that SL 82.0715 did not produce any neuronal vacuolization in the posterior cingulate and retrosplenial cortices (Fig. 1f). The ultrastructure of the neurons was very similar to that found in control rats. Cytoplasmic organelles such as endoplasmic reticulum, polyribosomes, mitochondria, microtubules and Golgi apparatus could be easily distinguished and were not altered. As previously reported [2, 16], an important cytoplasmic vacuolization of some retrosplenial and posterior cingulate neurons was observed in the present study after
6-Fig. 1. Comparative effects of an acute administration of dizocilpine and SL 82.0715 on neuronal morphology in the rat posterior cingulate cortex. a: semi-thin section of normal neurons in a control animal in layer III of the posterior cingulate cortex (arrows). The cytoplasm of large bipolar or pyramidal neurons is well stained. Bar = 30/2m. b: semi-thin section of vacuolated neurons (arrows) in an animal treated with 5 m g . k g ~ s.c. of dizocilpine. The cytoplasm of the altered neurons appears swollen and pale with n u m e r o u s vacuoles. Bar = 30/2m. c: semi-thin section of unaltered neurons (arrows) in an animal treated with 30 m g . kg ~ i.p. of SL 82.0715. The cytoplasm of the neurons is normally well stained. Bar = 30/2m. d: electron micrograph of a neuron in a control animal showing mitochondria (M), endoplasmic reticulum (ER), microtubules (MT), polyribosomes (P) and Golgi apparatus (G). Bar = 0.6/tm. e: electron micrograph of a neuron in an animal treated with dizocilpine (5 m g . kg -~, s.c.) showing swollen mitochondria (arrows), dilated endoplasmic reticulum (arrowheads), and phagolysosome (curved arrow). Note the normal mitochondria (M) in the adjacent tissue. Bar = 0.6/.tm. f: electron micrograph from an animal treated with 30 m g . kg -t i.p. of SL 82.0715 showing normal ultrastructure. Bar = 0.6/lm.
196
acute administration of 5 mg. kg ~ of dizocilpine, a dose which represents 25 times its neuroprotective dose in a model of focal ischemia in the mouse [10]. Although only light microscopic evaluation has been performed at the lower dose of dizocilpine (1 mg. kg -~, i.e. 5 times the neuroprotective dose in the mouse), pathomorphological changes could also be evidenced in 2 out of 3 animals. Pathomorphological changes of cerebrocortical neurons have also been reported with phencyclidine and two other N M D A channel ligands, tiletamine and ketamine in the rat. The order of potency with which these agents induced this effect (dizocilpine > phencyclidine > tiletamine > ketamine) is the same as their order of affinity for binding to the N M D A receptor channel site [16]. Moreover, an acute neuronal vacuolization has been found after local injection of the competitive N M D A receptor antagonist, D-2-amino-5-phosphonopentanoate, directly into the cingulate cortical region [16]. In contrast, the present study demonstrates that the polyamine site-directed N M D A antagonist SL 82.0715, even when administered at a dose of 30 mg. kg ~which represents 27 times its neuroprotective dose (EDs0 = 1.1 mg/kg, i.p.) in a model of focal cerebral ischaemia in the mouse [10], did not induce any neuropathological alteration in the posterior cingulate and retrosplenial cortices in the rat. Similarly, L-687,414 (the (+)cis-methyl analogue of HA 966), an N M D A receptor antagonist acting at the glycine site, also failed to alter cortical neuronal morphology [11]. These results suggest that although all types of N M D A receptor antagonists produce a blockade of N M D A receptor function, they have markedly different effects on cerebrocortical neuronal morphology. Thus, the appearance of morphological alterations in cerebrocortical neurons in response to N M D A receptor antagonists may depend on the specific component of the N M D A receptor with which they interact. The reason why N M D A receptor channel blockers but not polyamine and glycine site antagonists induce neurotoxic effects is as yet unclear. Different conductance levels or states of N M D A receptor-gated calcium channels may be affected by these drugs resulting in a qualitatively different blockade of N M D A receptor function. Alternatively, it is possible that the pathomorphological effects of N M D A receptor channel blockers in some cortical areas is unrelated to their interaction with the N M D A receptor, but is due to an unknown property of these drugs. The mechanism whereby N M D A receptor channel antagonists induce neuronal vacuolization of cortical neurons remains unclear. The presence of early degenerative changes such as cytoplasmic vacuoles may be the structural consequence of excessive cell metabolism in this region due to the large increase in cerebral metabolic rate
measured after the administration of N M D A receptorgated channel blockers [14]. Furthermore, after administration of dizocilpine or ketamine, the increase in cerebral blood flow in the posterior cingulate cortex greatly exceeds the increase in metabolism, thus leading to an 'uncoupling' between flow and metabolism [6, 15]. It is not clear whether the increased metabolic rate occurs in the same cells that manifest pathomorphological changes, or in other neural elements in the same region, but it seems likely that the two phenomena are causally related. In support of this view, L-687,414 and SL 82.0715 which do not increase limbic glucose metabolism in the rat brain [11, 19] do not induce pathomorphological changes in cingulate and retrosplenial cortical neurons. In conclusion, these findings suggest that neuroprotection mediated through blockade of the N M D A receptor complex is not necessarily associated with abnormal cerebrocortical neuronal morphology. The presence or absence of pathomorphological alterations with N M D A receptor antagonists seems to be related to the specific site of the N M D A receptor with which they interact. 1 Albers, G.W., Goldberg, M.R and Choi, D.W., N-Methyi-Daspartate antagonists: ready for clinical trial in brain ischaemia? Ann. Neurol., 25 (1989) 398~,03. 2 Allen, H.L. and Iversen, L.L., Phencyclidine, dizocilpine and cerebrocortical neurons, Science, 247 (1990) 221. 3 Balster, R.L., The behavioral pharmacology of phencyclidine. In H.Y. Meltzer (Ed.), Psychopharmacology: the Third Generation of Progress, Raven, New York, 1987, pp. 1573-1579. 4 Carter, C.J., Benavides, J., Legendre, P., Vincent, J.D., Noel, F., Thuret, F., Lloyd, K.G., Arbilla, S., Zivkovic, B., MacKenzie, E.T., Scatton, B. and Langer, S.Z., Ifenprodil and SL 82.0715 as cerebral anti-ischemic agents. II. Evidence for N-methyl-D-aspartate receptor antagonist properties, J. Pharmacol. Exp. Ther., 247 (1988) 1222 1232. 5 Carter, C.J., Lloyd, K.G., Zivkovic, B. and Scatton, B., Ifenprodil and SL 82.0715 as cerebral anti-ischaemic agents. III. Evidence for antagonistic effects at the polyamine modulatory site within the NMDA receptor complex, J. Pharmacol. Exp. Ther., 253 (1990) 475 482. 6 Cavazzuti, M., Porro, C.A., Biral, G.P., Benassi, C. and Barbieri, G.C., Ketamine effects on local cerebral blood flow and metabolism in the rat, J. Cereb. Blood Flow Metab., 7 (1987) 806-811. 7 Faden, A.I., Demediuk, R, Panter, S.S. and Vink, R., The role of excitatory amino acids and NMDA receptors in traumatic brain injury, Science, 244 (1989) 798 800. 8 France, C.P., Moerschbaecber, J.M. and Woods, J.M., MK-801 and related compounds in monkeys: discriminative stimulus effects and effects on a conditional discrimination, J. Pharmacol. Exp. Ther., 257 (1991) 727-734. 8 Gomez-Pinilla, F., Tram, H., Cotman, C.W. and Nieto-Sampedro, M., Neuroprotective effect of MK-801 and U-50488 H after contusive spinal cord injury, Exp. Neurol., 104 (1989) 118-124. 10 Goni, B., Benavides, J., MacKenzie, E.T. and Scatton, B., The pharmacotherapy of focal cortical ischaemia in the mouse, Brain Res., 522 (1990) 290 307.
197 11 Hargreaves, R.J., Rigby, M., Smith, D., Foster, A., Hurley, C.J. and Hill, R.G., L 687,414(+)cis-4-methyl-HA-966, an NMDA receptor antagonist acting at the glycine site, does not alter glucose metabolism or neuronal morphology at neuroprotective dose levels, J. Cereb. Blood Flow Metab., 11 Suppl. 2 (1991) S 301. 12 Jackson, A. and Sanger, D.J., Is the discriminative stimulus produced by phencyclidine due to an interaction with N-methyl-Daspartate receptors? Psychopharmacology, 96 (1988) 87-92. 13 Meldrum, B., Protection against ischaemic neuronal damage by drug acting on excitatory neurotransmission, Cerebrovasc. Brain Metab. Rev., 2 (1990) 27-57. 14 Nehls, D.G., Kurumaji, A., Park, C.K. and McCulloch, J., Differential effects of competitive and non-competitive N-methyl-Daspartate antagonists on glucose use in the limbic system, Neurosci. Lett., 91 (1988) 204-210. 15 Nehls, D.G., Park, C.K., MacCormack, A.G. and McCulloch, J., The effects of N-methyl-D-aspartate receptor blockade with MK801 upon the relationship between cerebral blood flow and glucose utilisation, Brain Res., 511 (1990) 271-279. 16 Olney, J.W., Labruyere, J. and Price, M.T., Pathological changes induced in cerebrocortical neurons by phencyclidine and related drugs, Science, 244 (1989) 1360-1362. 17 Perrault, G., Morel, E., Joly, D., Sanger, D.J. and Zivkovic, B., Comparison of the pharmacological profiles of four NMDA antagonists: ifenprodil, SL 82.0715, MK-801 and CPP in mice, Br. J. Pharmacol., 97 (1988) 580 P.
18 Scatton, B., Carter, C., Benavid~s, J. and Giroux, C., N-Methyl-Daspartate receptor antagonists: a novel therapeutic perspective for the treatment of ischemic brain injury, Cerebrovasc. Dis., 1 (1991) 121-135. 19 Scatton, B., Cudennec, A., Duverger, D., MacKenzie, E.T., Nowicki, J.P. and Zivkovic, B., Effects of SL 82.0715, an NMDA receptor antagonist acting at the polyamine modulatory site, on local cerebral glucose use in the rat brain, J. Cereb. Blood Flow Metab., 11 Suppl. 2 (1991) S 312. 20 Singh, L., Menzies, R. and Tricklebank, M.D., The discriminative stimulus properties of (+)HA-966, an antagonist at the glycine/Nmethyl-D-aspartate receptor, Eur. J. Pharmacol., 186 (1990) 129132. 21 Toulmond, S., Serrano, A., Benavides, J. and Scatton, B., Neuroprotective effects of SL 82.0715 in an experimental model of brain trauma, J. Cereb. Blood Flow Metab., 11 Suppl. 2 (1991) S 313. 22 Tricklebank, M.D., Singh, L., Oles, R.J., Preston, C. and Iversen, S.D., The behavioral effects of MK-801: a comparison with antagonists acting non-competitively and competitively at the NMDA receptor, Eur. J. Pharmacol., 167 (1989) 127-135. 23 Willetts, J., Balster, R.L. and Leander, J.D., The behavioral pharmacology of NMDA receptor antagonists, Trends Neurosci., 11 (1990) 423428.